Aust. Journal oj Social Issues
1979
John Braithwaite
and David Biles
Volume 14 Number 3
On Being Unemployed
and Being a Victim of
Crime
The results of the first Australian National Crime Victims Survey concerning the socio-economic status of victims of crime in Australia are
discussed. While the findings on occupational status and household income
are somewhat equivocal, the data show a strikingly higher rate of criminal
victimization among the unemployed tor a number of types of crime. The
unemployed, spending a large proportion of their lives in public rather than
private space, may paradoxically be more likely to be both victimized by
criminals and victimized by the police.
While there is considerable dispute in the criminological literature
over whether crime rates go up historically during periods of high unemployment, there is little dispute that at anyone point of time unemployed people are more likely than people with a job to commit
traditional criminal offences such as homicide, rape, robbery, assault,
theft, and breaking and entering.1
There is a long tradition of radical criminology which advances the
argument that capitalism creates crime by the brutalization of the
proletariat and the unemployed, and that the bourgeoisie then act to
protect themselves from the crime of the embittered proletariat by engaging police forces and building prisons. In more recent times it has
been suggested that the community action programs of the United
States War on Poverty of the 1960s were motivated by the fear of
middle class America at rising crime perpetrated by the largely black
unemployed. Implicit in so much of this speculation is the assumption
that it is the middle class who have most to fear from the crime of the
proletariat. Such an assumption can be traced back to the first radical
criminologist, Frederick Engels, in his work on crime and the condition
of the working class in England:
There is therefore no cause for surprise if the workers, treated as
brutes, actually become such; or if they maintain their consciousJohn Braithwaite-is a Criminologist, Australian Institute of Criminology, Canberra. David Biles is Assistant Director (Research), Australian Institute of
Criminology, Canberra.
192
UNEMPLOYED AS VICTIMS OF CRIMES
193
l1ess of manhood only by cherishing the most glowing hatred, the
most unbroken inward rebellion against the bourgeoisie in power. 2
In the present paper the results of the National Crime Victims Survey
conducted by the Australian Bureau of Statistics in 1975 are used to
establish whether it is in fact the middle class who have most to fear
from criminal victimization, and, more specifically, to compare the exten~ to which the unemployed become victims of crime with the victimization rate for people with jobs.
METHOD
The Sample
Dwellings for inclusion in the stratified multi-stage area sample were
selected from all parts of Australia excluding the Northern Territory,
rural regions, and locations with a population of less than 500 people.
Of 10,500 dwelling sites originally selected, 9,200 contained effective
households, of which 8,414 provided data for the survey. These households contained 18,694 persons ~ged 15 years and over, each of whom
supplied some data. The remarkabe household response rate of 91.5 %
was only possible, of course, in a survey which has the legal authority
of the Australian Bureau of Statistics.
The Crimes
Interview data were gathered on all victimizations during the previous
12 months for ten types of crime:
• Break and enter: breaking and entering a dwelling and then committing or intending to commit a crime in that dwelling.
• Motor vehicle theft: stealing or illegally using a motor vehicle or
using a motor vehicle without authorization.
• Robbery with violence: stealing which involves the threat or use of
actual violen'ce or force to a person or property.
• Theft: stealing without threatening or using violence or force to any
person or property.
• Fraud, forgery, false pretences: all types of fraud, forgery, uttering
(circulating any fraudulent document or money), falsification of records,
false pretences and all offences involving false claims, deception,
trickery, cheating or breaches of trust.
• Peeping: only females were asked if they had been spied upon by a
'peeping Tom'.
• Indecent exposure: only females were asked if a male had 'indecently
exposed' himself in front of them.
194
JOHN BRAITHWAITE AND DAVID BILES
• Rape and attempted rape: all rape, attempted rape and assault with
intent torape. Only females were asked about rape victimization.
• Nuisance calls: threats, abuses, indecent calls and other nuisance
calls by telephone.
• Assault: unlawful attack by one person on another for the purp.ose of
inflicting bodily injury.
RESULTS AND DISCUSSION
Standard Error
With a sample of such magnitude, problems of statistical inference
loom less large than with most social science data. Nevertheless, with
less common types of crime marginals can bec.ome quite small. As a
matter of policy the Bureau of Statistics will not make available raw
data on the number of actual victimizations of each type within the
sample. Instead we are provided with estimates weighted from the
sample for the number of victimizations nationally. There can be no
doubt that the Bureau's weighted national estimate is a superior statistic
to the raw figure. The weighting procedure is such that raw figures from
different geographical areas will be multiplied by different weights
depending on the proportion of the populati.on .of the nation living in
that area, and the response rate.
While the weighting procedure provides a superior statistic it does
create some complexity for the social scientist who might be interested
in calculating aconventi.onal test of statistical significance. Tests of significance have not been calculated for each comparison made in this
paper. However, Table 1 provides the standard errors for survey estimates of the number of victimizations of each type. s
It can be seen from Table 1 that the survey estimate of the number of
breaking and entering victimizations .occurring in Australia during 1975
was 146,500. The approximate percentage standard error on this estimate is 8.5. This means that the standard error is 8.5% of 146,500,
i.e., 12,500. Discounting non-sampling errors, there are therefore about
two chances in three that the number of breaking and enterings in Australia during 1975 fell between 1~4,000 and 159,000; and about 19
chances in 20 that it fell between 121,500 and 171,500.
Occupation and Crime Victimization
Occupation is the most widely used index of social class in the sociological literature. In Table 2 the ANU code for classifying occupations 4
has been applied to each respondent in the survey who was in the labour
195
UNEMPLOYED AS VICTIMS OF CRIMES
force at the time of interview. The first row of that table shows that the
highest rates .of break and enter victimization are found for the two
highest status occupational groups-professional and managerial. It is
possible that this simply reflects the fact that professional and managerial respondents were more likely to be designated as heads of households by the interviewers and therefore more likely to become eligible
for victimization for breaking and entering. Working wives and working
young people living with their parents are both less likely to be professionals or managers, and less likely to be designated as heads of
households.
TABLE 1
APPROXIMATE STANDARD ERROR PER CENT FOR SURVEY
ESTIMATES OF NUMBERS OF VICTIMIZATIONS IN AUSTRALIA FOR
1975, AND NUMBERS OF NON-REPORTED VICTIMIZATIONS IN
AUSTRALIA FOR 1975 BY TYPE OF CRIME
Assault
Robbery with violence
Rape, attempted rape
Break and enter
Motor vehicle theft
Fraud, forgery, false
pretences
Theft
Nuisance calls
*
Estimated
Number of
Victimizations
Standard
Error
Per Cent
Estimated
Number of
Non-Reported
Victimizations
Standard
Error
Per Cent
191,500
14,200
7,800
146,500
62,700
13.6
18.6
26.5
8.5
9.8
131,200
5,100
4,500
42,400
10,700
16.3
27.9
33.3
19.3
20.6
214,100
609,900
1,612,594
8.6
3.4
11.3
163,000
382,600
232,500*
9.9
4.2
23.6
The estimated number of non-reported victimizations relates only to the most
recent instance of a nuisance call.
TABLE 2
VICTIMIZATION RATES PER 100,000 POPULATION 15 YEARS AND OVER
BY OCCUPATION
Managers,
Farmers,
Sales,
Miners,
Shop
operators, labourers,
ProProtransport service
fessionals prietors Clerical Craftsmen workers workers
Break and enter
Motor vehicle theft
Ro b bery with violence
Theft
Fraud, forgery, false
pretences
Peeping
Indecent exposure
Rape, attempted rape
Nuisance calls
Assault
3,693.5 5,481.3 2,117.0 2,935.5
1,292.3 1,190.0 1,090.3 1,992.9
301.3
511.4
459.9
261.8
13,285.3 18,033.7 13,145.0 13,294.3
2,733.7 2,167.8
989.3 1,090.6
120.6
161.5
8,764.5 10,399.0
3,422.6 10,588.2 4,498.1 4,989.7 3,037.1 2,435.4
1,677.2 1,704.2 6,194.8
685.9 1,125.1 2,648.1
1,452.4
631.2
371.9
200.3
218.8
669.5
175.3
78.7
267.6
151.2
57.5
174.4
42,468.0 50,786.8 39,923.2 23,849.5 22,367.1 25,395.4
3,197.7 1,682.2 3,138.1 5,736.7 2,671.5 5,942.0
196
JOHN BRAITHWAITE AND DAVID BILES
Motor vehicle theft shows little variation by occupation, with craftsmen (skilled workers) having the highest rate of victimization. Standard
error on the robbery with violence estimates is high, so extreme caution
must be exercised in comparing these rates, but managerial and clerical
workers are more likely than others to report robbery victimizations.
Theft victimization does show a clear pattern by occupational status.
The two lowest status groups, 'sales, operators, transport workers' and
'mfners, labourers and service workers' (almost entirely unskilled
workers) are the groups with clearly the lowest victimization rates for
theft. Managers, farmers and shop proprietors are the group most likely
to report being the victim of theft. The same pattern applies with fraud,
forgery, false pretences victimization: managers have thy highest rates
and the two unskilled categories the lowest rates.
The over-representation of women in the clerical category explains
the "fact that it is clerical workers who report the highest victimization
rates ifor peeping. It is professionals who show the highest rates for being
victims of indecent exposure. Standard error on the rape estimates is too
high to permit meaningful comparisons. Again it is the two highest status
groups, professionals and managers, who are most likely to be victimized
by nuisance calls, possibly because they have the highest levels of access
toO telephones.
For assault the highest level of victimization is reported by 'miners,
labourers and service workers' and the second highest by craftsmen. This
latter finding suggests that one possible way to summarize the results of
Table 2 would be that while higher occupational status groups tend more
often toO be the victims of crimes against poverty, it may be the lowest
occupational groups who are most often the victims of some crimes
against the person. Before rushing to such a conclusion, however, we
should explore the effects nf another index of class position-household
income.
Average Weekly Household Income and Victimization
The income of all members of the household from a wide range of
sources over the previous twelve months was averaged to a weekly income according to a Bureau of Statistics formula. Included among the
forms of income which are eligible to be counted as part of the household incnme are gross salary or wages, superannuation payments,
worker's compensation, maintenance/alimony, rent (less expenses),
child endowment, and all forms of pensions, government benefits,
scholarships or student allowances.
The results in Table 3 suggest that the finding from the previous
section that respondents in prnfessional and managerial occupations are
UNEMPLOYED AS VICTIMS OF CRIMES
197
more likely to be victims of breaking and entering probably is a consequence of the fact that 'household heads' are more likely to be in professional and managerial occupations than working wives and children;
Table 3 shows that it is the households with the highest average weekly
incomes which are least likely to be victims of breaking and entering. In
contrast, these more affluent households are the most likely to be victims
of vehicle theft, presumably because they own more cars. Standard
error on the robbery with violence and rape estimates is really to.o large
to be able to make anything of the relatively small differences among
household income categories for these crimes.
TABLE 3
VICTIMIZATION RATES PER 100,000 POPULATION 15 YEARS AND OVER
BY GROSS WEEKLY INCOME OF HOUSEHOLD
Break and enter
Motor vehicle theft
Robbery with violence
Theft
Fraud, forgery, false
pretences
Peeping
Indecent exposure
Rape, attempted rape
Nuisance calls
Assault
$140-
$200-
$260-
<$80
$80-$139
$199
$259
$339
2,235.3
297.5
142.1
5,319.6
2,448.8
446.7
282.5
7,837.4
2,244.3
684.3
97.2
7,611.6
1,290.9 1,550.0
956.2
773.2
164.4
326.7
8,205.1 10,497.8
1,250.7
2,038.1
308.6
39.0
18,989.8
790.3
>$340
1,280.2
993.8
317.3
8,970.8
3,281.8
3,738.5 2,507.4 2,073.1 4,014.5
715.0 1,902.2 1,117.8 1,319.3 3,618.2
276.2
251.0
308.2
145.5
360.3
154.2
59.8
50.8
28.0
57.6
7,548.7 18,972.3 22,115.2 18,759.8 28,621.5
1,220.8 4,368.7 1,417.9 4,911.2 1,660.4
The correlation between household income and theft is less than perfect, but there is, nevertheless, a strong tendency for higher household
income to be associated with greater theft victimization. The positive
correlation with income is even stronger for fraud, forgery and false
pretences.
For both peeping and indecent exposure the people with the highest
victimization rates are those who live in both the very poorest and the
most affluent households. It is the middle income household which· reports the lowest levels of victimization for these minor sexual offences.
Both nuisance calls and assault show inconsistent relationships with
household income. The low assault rate of the lowest income group
possibly reflects the disproportionate number of aged pensioners in this
group. Certainly the household income data do not provide any glimmer
of a replication of the negative association between social class and
assault victimization reported in Table 2.
To summarise the results from both Tables 2 and 3, the National
Crime Victims Survey data on occupation and income indicate that it is
198
JOHN BRAITHWAITE AND DAVID BILES
the wealthy who are most likely to be victims of vehicle theft, fraud,
forgery, false pretences and common theft. The household income data
show that it is the poor who are most likely to be the victims of breaking
and entering. All other offence types tend to show inconsistent relationships with occupation and income.
Unemployment and Crime Victimization
Table 4 shows that the unemployed are a quite unique category with
respect to criminal victimization. The data indicated that the unemployed
are peculiarly vulnerable to being victims of some of the more serious
types of crime. Take theft for example. We have shown above how it is
generally true that the wealthy are more likely to be victims of theft
than the middle and low income earner. The unemployed, however,
have a far higher rate of theft victimization than both the employed
generally, and everyone of the household income groups listed in Table
3. The unemployed also show higher rates of criminal victimization than
the employed for breaking and entering, peeping, and assault. The
assault victimization rate is staggeringly high among the unemployed.
They are more than twice as likely to report victimization for this offence
than those in full-time jobs, and six times as likely to have been assaulted
compared with respondents not in the workforce or in part-time jobs.
TABLE 4
VICJ1IMIZATION RATES PER 100,000 POPULATION
15 YEARS AND OVER BY EMPLOY:MENT
Not in
workforce
Break and enter
Motor vehicle theft
Robbery with violence
Theft
Fraud, forgery, false
pretences
Peeping
Indecent exposure
Rape, attempted rape
Nuisance calls
AssauU
Unemployed
Full time
employment
Part time
employment
918.4
192.9
82.9
4,799.8
3,162.3
409.9
364.4
12,927.5
2,748.3
1,317.8
257.0
9,451.8
1,150.6
706.3
146.3
7,741.3
633.9
1,535.8
371.5
116.6
2,443.2
1,211.7
2,864.7
11,365.0
321.8
4,364.4
1,389.6
286.0
72.0
17,834.7
3,283.0
2,659.1
1,047.1
372.9
147.2
26,835.3
1,467.6
15,266.6
8,374.8
The unemployed do have lower rates of victimization for motor
vehicle theft and nuisance calls, probably because they do not own cars
or telephones. They are also less likely to report being victims of fraud,
forgery and false pretences-an expected finding since it is people in
business who generally report this kind of crime. Standard error with
respect to robbery, indecent exposure and rape is too high for any
UNEMPLOYED AS VICTIMS OF CRIMES
199
statement to be made about the rates for these offences among the unemployed.
Comparison with United States Findings
Data from the United States National Victims Survey5 of 1975 reveal
a more or less comparable picture to the one painted by the Australian
data.
The larger United States survey has an acceptable standard error for
rape and robbery with violence, so that a more complete picture is
possible than with the Australian data. Among the unemployed in the
US, victim reported crime rates were higher for rape, robbery with
violence, assault and theft. Lower income groups report higher rates of
victimization than higher income groups for rape, robbery with violence,
and assault. For theft we see the same kind of reversal that appears in
the Australian data: it is higher income groups who report the higher
rates of victimization for theft. In addition, as in the Australian data,
higher income groups report higher motor vehicle theft victimization.
CONCLUSION
It would seem that whether it is the working class or the middle class
who have most to fear from crime depends to a considerable extent upon
what kind of crime one is talking about and what is meant by working
class. The data from the Australian Crime Victims Survey show that in
exploring the class basis of criminal victimization it is important to
distinguish the unemployed from the working class generally. The unemployed in Australia are an underclass who experience a level of
criminal victimization far in excess of that experienced by either the
wealthy or the employed manual worker.
The excessive extent to which the unemployed are victims of crime in
Australia (and the United States) may reflect the fact that the unemployed spend such a large proportion of their time in public space.
Ironically, the fact that the unemployed spend so much of their time in
public space-in trains rather than cars, streets and parks rather than
factories and offices, public bars rather than private clubs-probably
renders them more available for both criminal victimization and
accusations by the police that they have been the perpetrators of crimes.
The data reported in this study point up how simplistic is the 'radical'
critique which tells us that crime is a manifestation of working class
rebellion against the ruling class, a way that the workers hit back at the
bourgeoisie. Jock Young is one of the first to have foreseen the need for
a more sophisticated radical critique which is less at odds with the
.
evidence:
200
JOHN BRAITHWAITE AND DAVID BILES
For the working class does have a real stake in a genuine social
order, however much it may be that conservative 'law and order'
campaigns are a sham behind which particular interests advance
themselves, and proclaim themselves to be acting in the interests of
all. It is a simple fact that the majority of working-class crime is
intra- and not inter-class in its choice of target, area of activity and
distribution. Working-class people suffer from crime,confront daily
the experience of material desperation, undergo the ravages of disorganization and competitive individualism. The ideology that plays
on this-bourgeois ideology-contains an element of truth, and
touches on the genuine interests of the class-albdt in a distorted
fashion.'6
Acknowledgments
This paper was made possible by the generous assistance and cooperation given to the Australian Institute of Criminology by the staff
of the Australian Bureau of Statistics.
REFERENCES
1. See the debate on this point in the Proceedings of Institute of Criminology,
University of Sydney, Unemployment and Crime, June 1978. See also the review of evidence in Brai,thwaite, J., Inequality, Cfime, and Public Policy,
London and Boston: Routledge and Kegan Paul, 1979.
2. Engels, F., The Condition of the Working Class in England in 1844, Harmondsworth: Penguin, 1969, pp. 144-145.
3. More detailed information on both standard error and sampling for the survey
is available in Australian Bureau of Statistics, 1975 General Social Survey:
Crime Victims, 4105.0, Canberra: Australian Bureau of Statistics, 1979.
4. Broom, L., Duncan-Jones, R., Lancaster-Jones, F., and McDonnell, P., Investigating Social Mobility, Department of Sociology, Research School of Social
Sciences, Departmental Monograph No.1, Canberra: Australian National
University, 1977.
5. See Got1fredson, M. R., Hindelang, M. J., and Parisi, N. (eds.), Sourcebook
on Criminal Justice Statistics-1977, Law Enforcement Assistance Administration, Washington, DC: Department of Justice, 1978, Tables 3.13-3.15, 3.43.
6. Young, J., Working Class Criminology, in Taylor, r., Walton, P., and Young, J.
(eds.), Critical Criminology, London and Boston: Routledge and Kegan Paul,
1975, p. 79.
Copyright of Full Text rests with the original copyright owner and, except as
permitted under the Copyright Act 1968, copying this copyright material is
prohibited without the permission of the owner or its exclusive licensee or
agent or by way of a license from Copyright Agency Limited. For information
about such licences contact Copyright Agency Limited on (02) 93947600 (ph) or
(02) 93947601 (fax)